microsoft/semantic-kernel · error · ServiceInitializationError

Ollama embedding model ID is not set.

Error message

Ollama embedding model ID is not set.

What it means

Thrown by the OllamaTextEmbedding constructor when the resolved embedding_model_id is empty or None after merging the ai_model_id argument with the OLLAMA_EMBEDDING_MODEL_ID environment variable. Without a model identifier the service cannot call the Ollama /api/embeddings endpoint, so it aborts. Raised as ServiceInitializationError before any network call.

Source

Thrown at python/semantic_kernel/connectors/ai/ollama/services/ollama_text_embedding.py:69

            ai_model_id (Optional[str]): The model name. (Optional)
            host (Optional[str]): URL of the Ollama server, defaults to None and
                will use the default Ollama service address: http://127.0.0.1:11434. (Optional)
            client (Optional[AsyncClient]): A custom Ollama client to use for the service. (Optional)
            env_file_path (str | None): Use the environment settings file as a fallback to using env vars.
            env_file_encoding (str | None): The encoding of the environment settings file, defaults to 'utf-8'.
        """
        try:
            ollama_settings = OllamaSettings(
                embedding_model_id=ai_model_id,
                host=host,
                env_file_path=env_file_path,
                env_file_encoding=env_file_encoding,
            )
        except ValidationError as ex:
            raise ServiceInitializationError("Failed to create Ollama settings.", ex) from ex

        if not ollama_settings.embedding_model_id:
            raise ServiceInitializationError("Ollama embedding model ID is not set.")

        super().__init__(
            service_id=service_id or ollama_settings.embedding_model_id,
            ai_model_id=ollama_settings.embedding_model_id,
            client=client or AsyncClient(host=ollama_settings.host),
        )

    @override
    async def generate_embeddings(
        self,
        texts: list[str],
        settings: "PromptExecutionSettings | None" = None,
        **kwargs: Any,
    ) -> ndarray:
        raw_embeddings = await self.generate_raw_embeddings(texts, settings, **kwargs)
        return array(raw_embeddings)

    @override

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Pass the model name directly: OllamaTextEmbedding(ai_model_id='nomic-embed-text')
2. Set the env var: export OLLAMA_EMBEDDING_MODEL_ID=nomic-embed-text
3. Add OLLAMA_EMBEDDING_MODEL_ID to your .env and pass env_file_path
4. Ensure the embedding model is pulled: ollama pull nomic-embed-text

Example fix

// before
embedder = OllamaTextEmbedding()
// after
embedder = OllamaTextEmbedding(ai_model_id='nomic-embed-text')

import os

model_id = 'nomic-embed-text'
if not model_id and not os.environ.get('OLLAMA_EMBEDDING_MODEL_ID'):
    raise RuntimeError('OLLAMA_EMBEDDING_MODEL_ID is not set and no ai_model_id provided')

from semantic_kernel.connectors.ai.ollama import OllamaTextEmbedding
embedder = OllamaTextEmbedding(ai_model_id=model_id)

from semantic_kernel.exceptions.service_exceptions import ServiceInitializationError

try:
    embedder = OllamaTextEmbedding()
except ServiceInitializationError as e:
    if 'embedding model ID' in str(e):
        embedder = OllamaTextEmbedding(ai_model_id=os.environ['FALLBACK_EMBEDDING_MODEL'])
    else:
        raise

When it happens

Trigger: Constructing OllamaTextEmbedding() with ai_model_id omitted while OLLAMA_EMBEDDING_MODEL_ID is not set in the environment or .env file. Also triggered by passing ai_model_id=None explicitly.

Common situations: New project with no .env configured; using a shared OLLAMA_HOST but forgetting the model-id env var; env var name typo (OLLAMA_EMBEDDING_MODEL instead of OLLAMA_EMBEDDING_MODEL_ID); pulling a text model but not an embedding model (ollama pull nomic-embed-text).

Related errors

• The Vertex AI embedding model ID is required.
• Ollama text model ID is required.
• The Amazon Bedrock Text Embedding Model ID is missing.
• The Google AI Gemini model ID is required.
• The Google AI embedding model ID is required.

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/389aab10139690dd. Report an issue: GitHub.